-
HI,
Thank you for your code and for adding BioBert.
Tried to use your code with SciBert. For this, I created a code 3 to specify the model type in the input (just as you introduced 2 for BioBert),…
-
It would save us a lot of time to have a button for duplicating in-statement properties (and when adding this, let's also use the same solution for relations). Typical use: we are attaching the same p…
-
After our branch get rebased, the test case failed because the updated `TokenClassificationPipeline` has added `stride` as an optional arguments that create conflicts with how we are defining `stride`…
-
Hi, the dataset for Baidu's paper: "ARNOR: Attention Regularization based Noise Reduction for Distant Supervision Relation Classification" was originally stored at: https://github.com/PaddlePaddle/mod…
-
**Describe the benchmark KG dataset**:
A subset of http://rtw.ml.cmu.edu/rtw/
Besides normal triples, there are also explicit positive **and negative** labels for the triple classification task.…
mberr updated
3 years ago
-
## Actions
- [ ] Get an exhaustive list of all the relations that we want to extract, e.g.
- `(GENE, is-in, BRAIN_REGION)`
- `(GENE, is-in, CELL_COMPARTMENT)`
- ...
- [ ] Define number o…
-
Hello,
I trained a knowbert language model on some medical data, and want to fine-tune it on a relation classification task. However, I get different Micro/Macro F1 scores every time a fine-tune it…
-
Corner case 1:
The IPynb below is made in VSCode. But once imported into Codepod,
1. all source code is gone.
2. HTML is not properly rendered.
![image](https://github.com/RunVas/RunVas/ass…
-
This is meant to gather any ideas for potential baselines or enhanced approaches to tackle the shared task.
-
Is it possible to create references from one data entry to another?
For example, it's common for a weapon to increase the damage of a specific spell. This obviously can be done with something like …